Friday, July 8, 2011

Endocannabinoids, Fat, and Rats

There has been a flurry of rather interesting my-style Evolutionary Psychiatry papers out this week.  Tomorrow (I hope), I'll review some findings about probiotics, gut microbiota, and brain function (in rodents, alas), and there is also a paper Jamie sent me about how breakfast staple types are correlated with gray matter and cognitive function in healthy children (If you have but two breakfast choices and you are forced to rely on correlative data, should you feed your children bread or rice if you are playing it safe and interested in protecting their growing brains?  Guess.)  Jamie might blog that one first, actually.  Who knows.  Today, though, I'm looking at the paper Endocannabinoid signal in the gut controls dietary fat intake published in PNAS.   

First off, I've added two new blogs to my "Of Like Minds" listing on the right,  Chris Kresser's Healthy Skeptic, and Anastasia's Primalmeded.  I've been meaning to add Chris for some time, but never really got around to it.  He's terrific and thoughtful.  Anastasia is a mom and medical student in Australia.  I find her relatively new blog to be no-nonsense and refreshing.

Now, the endocannabinoid paper.  Well.  If you haven't seen them yet, I've covered endocannabinoids previously:

For some background in rat studies, I am far more familiar with the binge eating literature than the obesity literature.  All the bingeing literature I'm aware of has been done with omega 6 fats (usually Crisco, though trans-fat free Crisco was used for more recent studies) and grains and sugar, and the consistent findings are: Rats will binge on sugar or fat (omega 6).  Only the combination of sugar and fat resulted in weight gain (otherwise bingeing rats will make up for extra calories by spontaneously restricting at other times).  One study showed bingeing rats maintained their weight but increased fat mass bingeing on crisco (I think.  I have to recheck that one.  I'll check and edit later if I am misremembering.)  

The major distinction in this literature comparing rats to humans is that humans do not typically binge on fat alone, but are far more likely to binge on pasta, bread, sweet foods, chocolate, and salty snacks.  The bingeing literature may not have the same focus and may not be as important to the overall state of obesity as the straight-up obesity literature - but with respect to "munchies" and endocannabinoids, I think it is a fair backdrop.

Alrighty then.  In the new paper, researchers gave rats "sham" feedings of (okay, three guesses as to what they used as the fat?  Pasture butter?  Coconut oil?  Lard?  WRONG) corn oil, a mixed "nutritionally complete" diet, sugar solutions, and protein solutions.  How do you sham feed a rat?  Well, very unpleasantly.  You install an aluminum cannula into the stomach and let the liquid food drain out as you allow the rats to eat a liquid diet of the aforementioned macronutrients.  Mmmm.  A spoonful of corn oil.  Delicious.

Here is what the researchers think about eating fat:

Mammals have an adaptive advantage in seeking fat-rich foods, which are nutritionally essential but scarce in most natural habitats.  This innate preference can become maladaptive… when it is not limited by environmental constraints.  Indeed, the unrestricted availability of fatty foods, which characterizes diets of industrial societies, is considered to be a key contributing factor for obesity, diabetes, and cardiovascular disease.

Now a little review of endocannabinoids.  They are a happy little family of omega-6 derived bioactive molecules that bind to the cannabis receptors (CB1 and CB2).  These receptors are also activated by cannabis.  The two best studied endocannabinoids are 2-AG and anandamide.  

The researchers in this study were trying to further elucidate the mechanisms by which endocannabinoid feedback goes between the mouth, gut, and brain.  It has been found previously that CB1 receptors on the tongue modulate neural activity elicited by a sweet taste (1).  Neural signals from nutrients - including fats and sugars, are transmitted from the mouth to the brain via the cranial nerves (specifically V, VII, IX, and X).  The twelve pairs of cranial nerves shoot out directly from the brain rather than being shuffled through the spinal cord first.  Most of them predictably control stuff in the head and neck.  One of the cranial nerves, the vagus (X), takes a long trip down to the gut, and seems to be responsible for a lot of the brain regulation of the gut and digestive system, and carries the feedback between the two areas.  

So, what happened with the sham feeding of sugar solution, protein solution, corn oil, or a mixed liquid diet and brain and nerve activity in these rats?  Levels of the endocannabinoids 2-AG and anadamide were increased in the jejunum  (the middle part of the small intestine) of rats fed the corn oil or mixed liquid diet, but not in rats fed the protein or sugar solutions.  None of the solutions changed the jejunal content of oleoethanolamide, a fat-derived molecule that typically signals satiety and is usually released by the small intestine in response to the ingestion of fat.  

Interestingly, severing the vagus nerve in these rats stopped the production of the endocannabinoids when the rats were sham-fed the corn oil.  This would suggest that feedback from the brain is required to produce endocannabinoids in the small intestine in response to a corn oil signal from the small intestine.  

The researchers did some more complicated work studying different entities along the metabolic pathways of the endocannabinoids to figure out exactly how the endocannabinoid levels increased.  Sham-feeding of corn oil didn't affect the generation of 2-AG, but did seem to slow down the natural breakdown of 2-AG (which is typically done via hydrolysis, biochem nerds.)  As for anandamide, the other popular endocannabinoid, sham-fat feeding both increased its production and decreased its breakdown.

In another twist of the study, the researchers measured the amount of corn oil or regular chow the rats consumed while they infused rimonabant, a cannabis receptor inverse agonist (in simplistic terms, it blocks the ability of the endocannabinoids to activate the cannabis receptors) into the jejunum.  The rats ate much less corn oil when they had guts full of rimonabant, and to a lesser extent, less regular chow.  Thus the researchers conclude that eating fat increases a positive feedback mechanism causing increased ingestion of foods (especially more fat) via a endocannabinoid signal between the gut and the brain along the vagus nerve.  They were excited to try gut-specific cannabis receptor blockers that wouldn't have all the pesky anxiety, insomnia, and depression side effects that rimonabant (which is active in the brain) does as a treatment for obesity.  It is also known (in rats) that small-intestine levels of 2-AG and anandamide rise in repsonse to food deprivation and fall upon refeeding, suggesting they may signal energy balance and promote caloric intake.

And, last but not least, the discussion goes into details about how reward area dopamine stimulation is affected by feeding rats fat-rich foods.  I don't have time to chase down all those papers right now, but let's guess which fat was used in those studies as well…

My take?  Sure, play it safe. Don't eat spoonfuls of corn oil.  Especially if you are a rat.  But elucidating some of these pathways is definitely interesting.  If there is some reason we should be extrapolating the sham feeding results to whole, real food eating in humans, please explain.  It seems to me the finding that the normal small intestinal fat satiety feedback molecule, oleoethanolamide, was not elevated with sham-feeding of corn oil or the mixed diets or sugar or protein suggests this model is incomplete at best.  


  1. Hi Emily, great post!
    isn’t there a chance that “pesky anxiety, insomnia, and depression” were caused by gut CB receptors (i.e., gut-specific knockouts would still experience pesky anxiety, etc.; also gut-specific antagonists would still cause pesky anxiety, etc.)?
    Thanks, Bill

  2. Hi Emily. Allow me to now officially thank you for the add. Cannot believe that I was placed in such distinguished company. I enjoy your posts tremendously and now find myself more and more interested in mental health (something I couldn't imagine after my less-than-exciting psych rotation). I appreciate the support and the occasional cycling Twitter banter :)

  3. Bill - it is certainly possible. But chemists may have a better time designing a CB blocker that isn't absorbed and only works on gut lining than they will in finding one that doesn't make a bunch of people nuts by direct cerebral action.